VEED Dubbing - ComfyUI Built-in Node Documentation

VEED Dubbing translates the speech in a video into another language with VEED, cloning each speaker’s voice and syncing their lips to the translation. The clip is uploaded, the VEED job is polled, and the dubbed video comes back at 25 fps with the same duration as the new audio track. Use it to localise a finished clip without re-shooting or re-recording the voice.

Inputs

Parameter	Description	Data Type	Required	Range
`model`	Dubbing model. `"standard"` is cheaper and returns mono audio, with Latin American Spanish and Brazilian Portuguese; `"pro"` adds regional accents and stereo audio. Selecting a model reveals `video`, `target_language`, `source_language` and `seed`.	DYNAMIC_COMBO	Yes	`"standard"` `"pro"`

standard Inputs

These inputs appear when model is set to "standard".

Parameter	Description	Data Type	Required	Range
`video`	Video with a visible face and speech, up to 599 seconds. The output is as long as its audio track and plays at 25 fps.	VIDEO	Yes	Up to 599 seconds
`target_language`	Language to dub into, one of 85 languages. (default: `"English"`)	COMBO	Yes	85 languages
`source_language`	Language spoken in the video, or `"auto"` to detect it. (default: `"auto"`)	COMBO	Yes	`"auto"` 85 languages
`seed`	Seed controls whether the node should re-run; results are non-deterministic regardless of seed. (default: 42)	INT	Yes	0 to 2147483647

pro Inputs

These inputs appear when model is set to "pro".

Parameter	Description	Data Type	Required	Range
`video`	Video with a visible face and speech, up to 599 seconds. The output is as long as its audio track and plays at 25 fps.	VIDEO	Yes	Up to 599 seconds
`target_language`	Language or regional accent to dub into: 87 languages plus 14 accents such as `"English (UK)"`, `"Spanish (Mexico)"` and `"Portuguese (Brazil)"`. (default: `"English (US)"`)	COMBO	Yes	101 options
`source_language`	Language spoken in the video, or `"auto"` to detect it. (default: `"auto"`)	COMBO	Yes	`"auto"` 91 languages
`seed`	Seed controls whether the node should re-run; results are non-deterministic regardless of seed. (default: 42)	INT	Yes	0 to 2147483647

The node stops with an error when the video has no audio track, when the clip is longer than 599 seconds, or when the VEED job fails.

Outputs

Output Name	Description	Data Type
`VIDEO`	The dubbed video: the original footage with the translated speech and matched lip movement.	VIDEO

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): fb818ca510e20bccffde61dca98c457984b8a110976e56f2a513f6c7f4d80c80

​Inputs

​standard Inputs

​pro Inputs

​Outputs

Inputs

standard Inputs

pro Inputs

Outputs